Metabolomics Society Forum

Software => R => XCMS => Topic started by: Hans on February 17, 2014, 07:52:34 AM

Title: plotTIC vs plotEIC vs getEIC vs rawEIC
Post by: Hans on February 17, 2014, 07:52:34 AM
Dear all,

While using the xcms-package I came across about 4 different ways to produce TIC or EIC figures. The figures show different results. Presumably I use the wrong parameter settings, or misunderstand some of the concepts. The example below illustrates my problem:

Code: [Select]
library(xcms)    
library(faahKO)
cdfpath <- system.file("cdf", package = "faahKO")
cdffiles <- list.files(cdfpath, recursive = TRUE,full=T)
xr <- xcmsRaw(cdffiles[1])
xr

par(mfrow=c(4,1))
plotTIC(xr)
plotEIC(xr, m=c(200,600), rt=c(2501.4, 4499.8))
t1 <- getEIC(xr, mzrange=cbind(200,600), rtrange=cbind(2501.4, 4499.8))
t2 <- cbind(xr@scantime, rawEIC(xr,mzrange=range(xr@env$mz))$intensity)

plot(t1)
plot(t2, type = "l")

This piece of code leads to the attached figure.

[attachment=0:3lvtgvgp]question.jpeg[/attachment:3lvtgvgp]

Could you help me understand the different outcomes? All help will be very much appreciated!

Kind regards, Hans

[attachment deleted by admin]
Title: Re: plotTIC vs plotEIC vs getEIC vs rawEIC
Post by: Ralf on February 22, 2014, 12:24:33 PM
Hans,

only rawEIC() extracts the EIC using the raw data directly, all other EIC methods make use of the so called profile matrix, a binned version of the raw data. Depending on how you set m/z step size for this matrix (profstep parameter), the results can be slightly different.

Regards,
Ralf